X-ray imaging is a crucial diagnostic tool in medicine and biomedical engineering, but image quality is often compromised by noise and artifacts. Traditional denoising methods may overly smooth or remove important features, limiting diagnostic accuracy. We propose a machine learning approach to X-ray image denoising, leveraging deep neural networks to separate noise from signal. The method deployed, trained to learn on a large dataset of X-ray images, learns to remove noise while preserving image features. Results of the proposed model show significant improvement in image quality, measured by peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) at 38.45(dB) and 0.92 respectively and in comparison with traditional method’s , peak signal-to-noise ratio (PSNR) result shows 35.12(dB) and structural similarity index (SSIM) result shows 0.85 . Comparing the results with the state-of-art, the proposed model approach has potential to enhance diagnostic accuracy, reduce radiation doses, and support image-guided interventions. This work demonstrates the promise of machine learning in X-ray image denoising, enabling improved healthcare outcomes and research advancements.